'Smooth talking gadgets' not far off say researchers

Joint effort to improve speech recognition and synthesis
Publish date:

A group of British universities are aiming to push forward efforts to create better intelligent speech recognition systems which, they claim, will soon result in human-like interaction with computers.

Researchers from the universities of Cambridge, Edinburgh and Sheffield announced a five year £6.2 million project aimed at using advanced computing techniques that will be able to understand spoken language and synthesise natural-sounding voice.

The researchers said that technology may soon be available which can recognise the voices of individuals and could become familiar with the expressions, accent and vocabulary of individual speakers.

"This research could open the door to computer speech technology becoming commonplace throughout our lives - at home, at work, and in our leisure time," said University of Edinburgh Professor Steve Renals.

Besides a new class of chatty gadgets, uses of the new speech technology include the potential to create voice-controlled devices for the home, which could help older people stay independent.